Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Khi nguoi dung can ap dung nguyen tac tam ly vao marketing — cham cam xuc, tao khan cap, xay trust, thuyet phuc mua. Cung dung khi nguoi dung nhac 'tam ly khach hang', 'psychology marketing', 'scarcity', 'FOMO', 'social proof', 'Cialdini', 'thuyet phuc khach', 'trigger cam xuc'. Skill nay giai thich 7 nguyen tac tam ly + cach dung vao copy, landing page, email, ads — dieu chinh theo van hoa Vietnam.
.claude/skills/minhnv0807-16-marketing-psychology/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 72% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 86% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 70% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 71% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 16% | 0% |
Doc .agents/product-marketing-context.md neu co.
Nguyen tac: Nguoi ta cam thay nghia vu tra on khi nhan duoc gia tri truoc.
Ap dung VN:
Cau trigger: "Tang ban ... — khong can mua"
Ket qua do duoc: Tang lead 2-3x so voi direct sell.
Nguyen tac: Khi da cam ket nho, nguoi ta xu huong tiep tuc cam ket lon hon.
Ap dung VN:
Cau trigger: "Ban muon ... dung khong?" (khach tra loi co roi moi chao ban)
Nguyen tac: Nguoi ta tin hanh dong dam dong hon quang cao.
Ap dung VN (cuc quan trong o VN):
Cau trigger: "Con so] khach da ..." / "Ten nguoi] noi: ..."
Nguyen tac: Nguoi ta tin chuyen gia co uy tin.
Ap dung VN:
Cau trigger: "Ten chuyen gia] - chung chi/kinh nghiem]"
Nguyen tac: Nguoi ta mua tu nguoi ho thich.
Ap dung VN:
Cau trigger: "Chi ten] — founder - chia se: ..."
Nguyen tac: Cai gi khan hiem duoc cho la co gia tri hon.
Ap dung VN (can that, khong gia):
Luu ý: Khach VN da rat quen scarcity fake → chi dung khi that su khan hiem. Neu fake bi phat hien = mat trust manh.
Cau trigger: "Chi con so cu the] — deadline cu the]"
Nguyen tac: Nguoi ta mua tu brand/nhom ho thuoc ve.
Ap dung VN:
Cau trigger: "Neu ban la nhan dang] — bai nay danh cho ban"
| Tep | Ap dung uu tien | Tranh | |-----|----------------|-------| | Cold (TOFU) | Reciprocity (tang gia tri), Social proof (con so), Liking (story) | Scarcity (som qua), Authority gia | | Warm (MOFU) | Authority (chuyen mon), Social proof (review cu the), Commitment (micro-yes) | Scarcity (chua du tin de hieu la that) | | Hot (BOFU) | Scarcity (deadline that), Commitment (tiep tuc yes), Unity (nhom VIP) | Too much reciprocity (chan cong ty moi) |
[Hero] Authority: "[Ten chuyen gia] - [uy tin]" + Unity: "Danh cho [nhan dang]"
[Problem] Liking: "Chung minh hieu ban dang ..."
[Solution] Reciprocity: Tang checklist/demo free
[Proof] Social Proof: So khach + review + logo bao chi
[Price] Scarcity: "Chi [X] slot con lai tuan nay" (neu that)
[CTA] Commitment: "Ban sap trai nghiem ..." (tien luong)Curiosity (to mo): "Tai sao 94% khach spa khong quay lai lan 2?"
Authority (uy tin): "[Bac si da lieu]: 3 sai lam khi trj mun"
Scarcity (khan cap): "24h cuoi — uu dai [con so] khach dau"
Social proof (dam dong): "1,200 chi da chuyen sang ... — co ban chua?"
Personal (ca nhan): "[Ten khach], minh co gi do cho ban"[Social proof]: "1,247 khach spa tin dung — ban thu chua?"
[Scarcity]: "Chi 30 slot tuan nay — giam 30% facial premium"
[Authority]: "Bac si da lieu 15 nam - Chia se quy trinh tri mun khong de seo"
[Liking]: "Chao, minh la [Ten founder]. Minh tao spa nay vi ..."product-marketing-context)| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-11 | fail→pass | 13,376 | 8,922 | -33% | 1 | 1 | 0% | 2,038 | 3,501 | +72% | 0 | 0 | — |
case-01 | fail→fail | 28,696 | 26,811 | -7% | 1 | 1 | 0% | 4,062 | 5,881 | +45% | 0 | 0 | — |
case-02 | fail→fail | 26,054 | 32,944 | +26% | 1 | 1 | 0% | 4,005 | 7,035 | +76% | 0 | 0 | — |
case-03 | fail→fail | 16,608 | 16,793 | +1% | 1 | 1 | 0% | 2,598 | 4,689 | +80% | 0 | 0 | — |
case-04 | fail→pass | 17,667 | 15,521 | -12% | 1 | 1 | 0% | 2,447 | 4,548 | +86% | 0 | 0 | — |
case-05 | fail→fail | 20,242 | 19,104 | -6% | 1 | 1 | 0% | 2,839 | 4,813 | +70% | 0 | 0 | — |
case-06 | pass→pass | 14,662 | 15,905 | +8% | 1 | 1 | 0% | 2,279 | 4,426 | +94% | 0 | 0 | — |
case-07 | pass→pass | 16,188 | 16,870 | +4% | 1 | 1 | 0% | 2,417 | 4,782 | +98% | 0 | 0 | — |
case-08 | pass→pass | 15,080 | 12,290 | -19% | 1 | 1 | 0% | 2,344 | 3,946 | +68% | 0 | 0 | — |
case-09 | pass→pass | 16,600 | 12,944 | -22% | 1 | 1 | 0% | 2,553 | 4,138 | +62% | 0 | 0 | — |
case-10 | fail→pass | 16,970 | 15,033 | -11% | 1 | 1 | 0% | 2,619 | 4,458 | +70% | 0 | 0 | — |
case-12 | pass→pass | 14,053 | 12,580 | -10% | 1 | 1 | 0% | 2,402 | 4,174 | +74% | 0 | 0 | — |
case-13 | fail→pass | 15,581 | 13,457 | -14% | 1 | 1 | 0% | 2,567 | 4,402 | +71% | 0 | 0 | — |
case-14 | pass→pass | 18,500 | 15,837 | -14% | 1 | 1 | 0% | 2,634 | 4,414 | +68% | 0 | 0 | — |
case-15 | pass→pass | 14,465 | 16,784 | +16% | 1 | 1 | 0% | 2,321 | 4,537 | +95% | 0 | 0 | — |
case-16 | pass→pass | 16,239 | 15,759 | -3% | 1 | 1 | 0% | 2,659 | 4,698 | +77% | 0 | 0 | — |
case-17 | pass→pass | 12,237 | 15,764 | +29% | 1 | 1 | 0% | 1,841 | 4,568 | +148% | 0 | 0 | — |
case-18 | pass→pass | 18,657 | 17,135 | -8% | 1 | 1 | 0% | 2,892 | 4,867 | +68% | 0 | 0 | — |
case-19 | pass→pass | 12,661 | 11,893 | -6% | 1 | 1 | 0% | 1,964 | 4,042 | +106% | 0 | 0 | — |
case-20 | pass→pass | 9,877 | 8,380 | -15% | 1 | 1 | 0% | 1,588 | 3,281 | +107% | 0 | 0 | — |
case-21 | pass→pass | 10,498 | 5,565 | -47% | 1 | 1 | 0% | 1,665 | 3,034 | +82% | 0 | 0 | — |
case-22 | fail→pass | 13,230 | 3,021 | -77% | 1 | 1 | 0% | 2,192 | 2,533 | +16% | 0 | 0 | — |
case-23 | pass→pass | 16,195 | 10,220 | -37% | 1 | 1 | 0% | 2,785 | 3,852 | +38% | 0 | 0 | — |
case-24 | pass→pass | 17,107 | 17,554 | +3% | 1 | 1 | 0% | 2,696 | 4,973 | +84% | 0 | 0 | — |
case-25 | pass→pass | 15,962 | 18,242 | +14% | 1 | 1 | 0% | 2,811 | 5,211 | +85% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 25 cases were attempted. The headline lift of +20 percentage points is the difference between those two pass rates over the 25 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.